Thanks guys.
We are better with backups as we have split the dbs across different availability groups with diff power supply etc.
So it is expected that servers will crash and we need to failover the db onto the other regions avg if required.
but servers crash all the time and we have been asked to make arrangements thinking of that.
So as long as this is a h/w failure i guess there is only one db server getting impacted and we should be good which is
good
If this was a pg bug all the nodes would get impacted and we would be in deep waters :)
So thanks.
Regards,
Vijay
________________________________________
From: Rui DeSousa <rui@crazybean.net>
Sent: Thursday, June 28, 2018 8:13:16 PM
To: Vijaykumar Jain
Cc: Scott Ribe; pgsql-admin@lists.postgresql.org
Subject: Re: [External] Re: standby replication server throws invalid memory alloc request size , does not start up
> On Jun 28, 2018, at 9:47 AM, Vijaykumar Jain <vjain@opentable.com> wrote:
>
> Thanks Scott.
>
> fsync is on.
>
> postgres=# show fsync;
> fsync
> -------
> on
> (1 row)
>
Yeah, I agree with Scott; it sounds like fsync() is being shorted somewhere in the I/O path. Are you using a RAID
controller;if so is the cache setting — write through or write back? If it’s write back then writes are being cached
anda power outage will lead to data corruption — the only saving grace at that point would be the battery backup logic
onthe controller which may or nay not be trustworthy.